Sky blue or murky waters: The diagnostic utility of methylene blue

Abstract

See Articles, pp. 289, 294. The diagnosis of Barrett's esophagus has taken on new meaning with the recognition that the extent and location of specialized intestinal metaplasia (SIM) of the esophagus may have different prognostic implications.1Hirota WK Loughney TM Lazas DJ Maydonovitch CL Rholl V Wong KH. Specialized intestinal metaplasia, dysplasia, and cancer of the esophagus and esophagogastric junction: prevalence and clinical data.Gastroenterology. 1999; 116: 277-285Abstract Full Text Full Text PDF PubMed Scopus (462) Google Scholar, 2Weston AP Badr AS Hassanein RS. Prospective multivariate analysis of clinical, endoscopic, and histological factors predictive of the development of Barrett's multifocal high-grade dysplasia or adenocarcinoma.Am J Gastroenterol. 1999; 94: 3413-3419Crossref PubMed Google Scholar, 3Weston AP Krmpotich PT Cherian R Dixon A Topalosvki M. Prospective long-term endoscopic and histological follow-up of short segment Barrett's esophagus: comparison with traditional long segment Barrett's esophagus.Am J Gastroenterol. 1997; 92: 407-413PubMed Google Scholar, 4Rudolph RE Vaughan TL Storer BE Haggitt RC Rabinovitch PS Levine DS et al.Effect of segment length on risk for neoplastic progression in patients with Barrett's esophagus.Ann Intern Med. 2000; 132: 612-620Crossref PubMed Scopus (255) Google Scholar However, to clearly map the entire columnar epithelium is not as simple as long-segment Barrett's esophagus (LSBE) and, to a lesser degree, short-segment Barrett's esophagus (SSBE) is a mosaic of differing columnar epithelia that include junctional, cardiac, and SIM. The identification of SIM and especially dysplasia is of great clinical importance because of the potential for malignant transformation. Unfortunately, standard endoscopes allow us to view tissue macroscopically but not microscopically, which makes identification of differing histologic characteristics impossible. Techniques to overcome this limitation have been developed such as laser-induced fluorescence, Raman spectroscopy, optical coherence tomography, chromoendoscopy with dyes (methylene blue, crystal violet, congo red) alone or in combination and with or without magnification endoscopy. Economically, methylene blue (MB) staining is attractive and seems to be practical because the technique is relatively simple to learn and requires only acetylcysteine solution, methylene blue solution, water, and a washing catheter. Despite its simplicity, there are conflicting reports concerning the accuracy and practicality of methylene blue chromoendoscopy for the diagnosis of SIM and dysplasia. As more data are presented, one may even question whether this technique is better than the conventional technique of 4-quadrant random biopsies at 2-cm intervals. Two articles published in this issue of Gastrointestinal Endoscopy, by Wo et al5Wo JM Ray MB Mayfield-Stokes S Al-Sabbagh G Gebrail F Slone SP et al.Comparison of methylene blue-directed biopsies and conventional biopsies in the detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2001; 54: 294-301Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar and Sharma et al,6Sharma P Topalovski M Mayo MS Weston AP. Methylene blue chromoendoscopy for detection of short segment Barrett's esophagus.Gastrointest Endosc. 2001; 54: 289-293Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar are reflective of these differences. In that of Wo et al, 47 patients entered the study with 35 patients undergoing both MB staining and conventional 4-quadrant biopsies at 2 separate sessions. When undergoing MB staining, 4-quadrant biopsy specimens were obtained at 2-cm intervals of both the MB stained and unstained areas. If the intervening areas had either homogeneous or heterogeneous MB staining, 4-quadrant biopsy specimens were obtained at 1-cm intervals. Of the 35 patients, 15 had LSBE and 20 had SSBE. Nine patients with SSBE (19%) had no appreciable staining, 6 (13%) had diffuse staining, and 32 (68%) had nondiffuse staining. The frequency at which SIM was identified in all MB stained specimens versus conventional specimens was similar overall (73% vs. 71%), even when categorized in terms of LSBE (84% vs. 80%) or SSBE (41% vs. 51%) respectively. Dysplasia was noted in 20% of MB-directed (stained and unstained) versus 18% of conventional specimens. The proportion of specimens with dysplasia remained similar when only MB-stained (23%) were compared with MB-unstained (18%) biopsy specimens. Although there was a higher degree of dysplasia in patients with LSBE (22%-24%) versus those with SSBE (8%-10%), no difference in detection rate was noted between the MB-stained and the conventional biopsy specimens. Interestingly, MB staining was less likely to diagnose SIM than conventional biopsies in SSBE because 7 patients with SIM had no appreciable staining. The MB-staining technique required a significantly longer time than the conventional technique to perform (21 vs. 14 minutes, respectively; p < 0.01) and there were more specimens taken in the MB staining vs. the conventional group. Overall, the sensitivity and specificity of MB staining to detect SIM was poor at 56% and 51%, and for dysplasia 51% and 48%, respectively. In contradistinction to the study of Wo et al., Sharma et al.6Sharma P Topalovski M Mayo MS Weston AP. Methylene blue chromoendoscopy for detection of short segment Barrett's esophagus.Gastrointest Endosc. 2001; 54: 289-293Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar noted a strong correlation between MB staining and the detection of SIM in patients with SSBE. In this study, 75 patients with columnar mucosa emanating from the distal esophagus for up to 3 cm were evaluated with MB, whereas a different group of 83 patients with SSBE with similar endoscopic findings comprised the control population. In the MB-stained group, biopsy specimens were obtained in the stained and unstained areas, whereas in the conventional group, specimens were obtained in 4 quadrants at 2-cm intervals for circumferential mucosa or at least 2 specimens per centimeter for each columnar-lined tongue. Specialized intestinal metaplasia was noted in 61% of the MB group and 42% of the control population (p < 0.02). When the length of the Barrett's was less than 1 cm, there was no difference in the identification of SIM between the MB stained and conventionally stained group (17% vs. 25%). However, with increasing lengths of Barrett's epithelium, significantly more SIM was identified when MB staining was compared with conventional staining (77% vs. 45%, at 1-2 cm, p < 0.03 and 90% vs. 58%, at 2-3 cm, p < 0.02, respectively). With shorter lengths of columnar epithelium (0-2 cm), 6 of 75 cases were not stained with MB. Low-grade dysplasia was noted only in the MB-stained group (n = 6), and all dysplasia exhibited MB staining. Additionally, fewer specimens were required in the MB-stained group; 4.3 versus 5.1 in the conventional group (p < 0.01). These two studies exemplify the ongoing differences in the published studies concerning the application of MB staining to the identification of SIM and dysplasia. It seems that for every favorable study describing the use of methylene blue chromoendoscopy for Barrett's esophagus,6Sharma P Topalovski M Mayo MS Weston AP. Methylene blue chromoendoscopy for detection of short segment Barrett's esophagus.Gastrointest Endosc. 2001; 54: 289-293Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar, 7Canto MI Setrakian S Petras RE Blades E Chak A Sivak Jr., MV Methylene blue selectively stains intestinal metaplasia in Barrett's esophagus.Gastrointest Endosc. 1996; 44: 1-7Abstract Full Text Full Text PDF PubMed Scopus (270) Google Scholar, 8Canto MIF Setrakian S Willis J Chak A Petras R Powe NR et al.Methylene blue-directed biopsies improve detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2000; 51: 560-568Abstract Full Text Full Text PDF PubMed Scopus (257) Google Scholar, 9Kiesslich R Han M Herrmann G Jung M. Screening for specialized columnar epithelium with methylene blue: chromoendoscopy in patients with Barrett's esophagus and a normal control group.Gastrointest Endosc. 2001; 53: 47-52Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar there are other studies that reveal it as no better than conventional biopsies.5Wo JM Ray MB Mayfield-Stokes S Al-Sabbagh G Gebrail F Slone SP et al.Comparison of methylene blue-directed biopsies and conventional biopsies in the detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2001; 54: 294-301Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar, 10Dave U Shousha S Westaby D. Methylene blue staining: is it really useful in Barrett's esophagus?.Gastrointest Endosc. 2001; 53: 333-335Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar These and other conflicting reports may be related to several differences in the way the studies were conducted. The significant difference between the study of Wo et al.5Wo JM Ray MB Mayfield-Stokes S Al-Sabbagh G Gebrail F Slone SP et al.Comparison of methylene blue-directed biopsies and conventional biopsies in the detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2001; 54: 294-301Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar and that of Sharma et al.6Sharma P Topalovski M Mayo MS Weston AP. Methylene blue chromoendoscopy for detection of short segment Barrett's esophagus.Gastrointest Endosc. 2001; 54: 289-293Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar is the study design. Although Sharma et al. studied more patients, MB staining and conventional biopsy specimens were compared by using two different groups of patients that might have differed with respect to the distribution and extent of SIM in the esophagus. On the other hand, Wo et al.5Wo JM Ray MB Mayfield-Stokes S Al-Sabbagh G Gebrail F Slone SP et al.Comparison of methylene blue-directed biopsies and conventional biopsies in the detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2001; 54: 294-301Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar performed a randomized crossover design to compare the two techniques within the same group of patients and obtained specimens from both MB-positive stained areas and MB-negative stained areas. The MB biopsy protocol was systematic and was not specifically intended to answer the question of whether fewer MB-stained specimens resulted in a yield that is similar to that of random 4-quadrant biopsies. Also, Sharma et al. expressed the sensitivity of MB to detect SIM as a per-patient ratio, whereas Wo et al. expressed sensitivity as a per biopsy ratio, which provides a closer reflection of how well the MB technique identifies SIM. It is apparent that studies vary widely in terms of study population and sample size, the technique of applying methylene blue, the protocol for obtaining biopsy specimens, the staining pattern after MB application, and the way in which the data are expressed. All of these factors can affect the results. In a sense our purpose is to answer several important questions at the same time: “What tissue does MB stain and what is the optimal staining technique?” and “what is the diagnostic utility of MB in detecting SIM and dysplasia?” The answer to the first question impacts to a significant extent the answer to the second question. Since the inception of methylene blue chromoendoscopy in the esophagus as a means of detecting SIM, the question of how and why this stain works has been fascinating. Considered a vital stain and not merely a tissue dye, the stain is said to be taken into the cytoplasm of actively absorbing cells found in the normal intestine, colon, and in both gastric and specialized intestinal metaplasia.11Canto MI. Vital staining and Barrett's esophagus.Gastrointest Endosc. 1999; 49: S12-S16Abstract Full Text Full Text PDF PubMed Scopus (49) Google Scholar, 12Carr-Locke DL Al-Chaws FH Branch MS Byrne WJ Edmundowicz SA Jamidar PA et al.Technology assessment status evaluation: endoscopic tissue staining and tattooing.Gastrointest Endosc. 1996; 43: 652-656Abstract Full Text Full Text PDF Scopus (9) Google Scholar Knowledge remains imperfect as to which cytoplasmic compartment of absorptive cells the stain is taken up into. There are early reports on the use of MB to stain intestinal metaplasia of the stomach; in an early study from Japan, pronase was used as the proteolytic (mucolytic) agent followed by 0.7% methylene blue.13Ida K Hashimoto Y Kawai K. In vivo staining of gastric mucosa: its application to endoscopic diagnosis of intestinal metaplasia.Endoscopy. 1975; 7: 18-24Crossref Scopus (48) Google Scholar The investigators described the process by which methylene blue was taken up into the cell based on its indirect incorporation into the cells along with a fat droplet rather than staining only the tissue surface. This study was used as the basis for subsequent studies in the United States in which acetylcysteine was substituted for pronase, the latter agent being unavailable in the United States.14Fennerty MB Sampliner RE McGee DL Hixson LJ Garewal HS. Intestinal metaplasia of the stomach: identification by a selective mucosal staining technique.Gastrointest Endosc. 1992; 38: 696-698Abstract Full Text PDF PubMed Scopus (73) Google Scholar This study was concerned with staining of gastric intestinal metaplasia, not esophageal SIM, and only described the process of staining without explanation as to how the stain worked or why it was specific for SIM. Personal experience with this stain ingested orally revealed it to be nonspecific and highly absorptive, staining the mouth and tongue for 24 hours. Four years later, Canto et al.7Canto MI Setrakian S Petras RE Blades E Chak A Sivak Jr., MV Methylene blue selectively stains intestinal metaplasia in Barrett's esophagus.Gastrointest Endosc. 1996; 44: 1-7Abstract Full Text Full Text PDF PubMed Scopus (270) Google Scholar published the first study in which a startling 95% of SIM, dysplasia, and cancer of the esophagus was detected by MB staining. The implications of this study were clear; in large segments of LSBE “blind eyes” were no longer used, but for the first time, areas of SIM, dysplasia, and cancer could be seen by using a simple technique, easily learned and performed at minimal costs. However, over the subsequent years, strong controversy has developed concerning this technique because numerous studies have not reproduced such excellent results. Among the 6 published studies concerning MB staining of SIM in the esophagus,5Wo JM Ray MB Mayfield-Stokes S Al-Sabbagh G Gebrail F Slone SP et al.Comparison of methylene blue-directed biopsies and conventional biopsies in the detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2001; 54: 294-301Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar, 6Sharma P Topalovski M Mayo MS Weston AP. Methylene blue chromoendoscopy for detection of short segment Barrett's esophagus.Gastrointest Endosc. 2001; 54: 289-293Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar, 7Canto MI Setrakian S Petras RE Blades E Chak A Sivak Jr., MV Methylene blue selectively stains intestinal metaplasia in Barrett's esophagus.Gastrointest Endosc. 1996; 44: 1-7Abstract Full Text Full Text PDF PubMed Scopus (270) Google Scholar, 8Canto MIF Setrakian S Willis J Chak A Petras R Powe NR et al.Methylene blue-directed biopsies improve detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2000; 51: 560-568Abstract Full Text Full Text PDF PubMed Scopus (257) Google Scholar, 9Kiesslich R Han M Herrmann G Jung M. Screening for specialized columnar epithelium with methylene blue: chromoendoscopy in patients with Barrett's esophagus and a normal control group.Gastrointest Endosc. 2001; 53: 47-52Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar, 10Dave U Shousha S Westaby D. Methylene blue staining: is it really useful in Barrett's esophagus?.Gastrointest Endosc. 2001; 53: 333-335Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar 5 included patients with both LSBE and SSBE; 1 study investigated only patients with SSBE.6Sharma P Topalovski M Mayo MS Weston AP. Methylene blue chromoendoscopy for detection of short segment Barrett's esophagus.Gastrointest Endosc. 2001; 54: 289-293Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar The number of patients varied in each group with respect to LSBE (n = 9-35) and SSBE (n = 8-75). Biopsy specimens were obtained a second time in the same patients on a separate occasion by using either the MB or random 4-quadrant biopsy technique in only 2 studies.5Wo JM Ray MB Mayfield-Stokes S Al-Sabbagh G Gebrail F Slone SP et al.Comparison of methylene blue-directed biopsies and conventional biopsies in the detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2001; 54: 294-301Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar, 8Canto MIF Setrakian S Willis J Chak A Petras R Powe NR et al.Methylene blue-directed biopsies improve detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2000; 51: 560-568Abstract Full Text Full Text PDF PubMed Scopus (257) Google Scholar In one study, biopsy specimens were obtained randomly after specimens of MB-stained areas were taken,9Kiesslich R Han M Herrmann G Jung M. Screening for specialized columnar epithelium with methylene blue: chromoendoscopy in patients with Barrett's esophagus and a normal control group.Gastrointest Endosc. 2001; 53: 47-52Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar whereas in 3 studies the control group consisted of an entirely separate group of patients with and without Barrett's esophagus.6Sharma P Topalovski M Mayo MS Weston AP. Methylene blue chromoendoscopy for detection of short segment Barrett's esophagus.Gastrointest Endosc. 2001; 54: 289-293Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar, 7Canto MI Setrakian S Petras RE Blades E Chak A Sivak Jr., MV Methylene blue selectively stains intestinal metaplasia in Barrett's esophagus.Gastrointest Endosc. 1996; 44: 1-7Abstract Full Text Full Text PDF PubMed Scopus (270) Google Scholar, 9Kiesslich R Han M Herrmann G Jung M. Screening for specialized columnar epithelium with methylene blue: chromoendoscopy in patients with Barrett's esophagus and a normal control group.Gastrointest Endosc. 2001; 53: 47-52Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar The MB staining pattern was reported in 3 studies and described as focal in 68% to 80% of patients in 2 studies (LSBE + SSBE)5Wo JM Ray MB Mayfield-Stokes S Al-Sabbagh G Gebrail F Slone SP et al.Comparison of methylene blue-directed biopsies and conventional biopsies in the detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2001; 54: 294-301Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar, 7Canto MI Setrakian S Petras RE Blades E Chak A Sivak Jr., MV Methylene blue selectively stains intestinal metaplasia in Barrett's esophagus.Gastrointest Endosc. 1996; 44: 1-7Abstract Full Text Full Text PDF PubMed Scopus (270) Google Scholar and diffuse in 77% of patients with SSBE in one study.5Wo JM Ray MB Mayfield-Stokes S Al-Sabbagh G Gebrail F Slone SP et al.Comparison of methylene blue-directed biopsies and conventional biopsies in the detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2001; 54: 294-301Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar Generally, focal staining was seen more commonly in LSBE versus SSBE. Eight percent to 20% of patients with Barrett's esophagus did not exhibit staining with MB in the 3 studies for which these data were provided. Although biopsy specimens were obtained of MB-stained areas in all studies, little information is provided concerning the protocol followed in obtaining tissue specimens. No information is given with respect to whether specimens were obtained from light, medium, or darkly stained areas. Wo et al.5Wo JM Ray MB Mayfield-Stokes S Al-Sabbagh G Gebrail F Slone SP et al.Comparison of methylene blue-directed biopsies and conventional biopsies in the detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2001; 54: 294-301Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar alone noted that specimens were obtained at 2-cm intervals of MB stained and unstained areas in addition to 4-quadrant biopsies of intervening homogeneously or heterogeneously stained areas. The mean number of specimens taken per patient varied from 3 to 4.3 in an SSBE group to 14 in a mixed group of LSBE and SSBE patients. Three studies noted a mean increase in time to perform MB staining of 7 to 8 minutes.5Wo JM Ray MB Mayfield-Stokes S Al-Sabbagh G Gebrail F Slone SP et al.Comparison of methylene blue-directed biopsies and conventional biopsies in the detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2001; 54: 294-301Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar, 8Canto MIF Setrakian S Willis J Chak A Petras R Powe NR et al.Methylene blue-directed biopsies improve detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2000; 51: 560-568Abstract Full Text Full Text PDF PubMed Scopus (257) Google Scholar, 10Dave U Shousha S Westaby D. Methylene blue staining: is it really useful in Barrett's esophagus?.Gastrointest Endosc. 2001; 53: 333-335Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar One group of investigators reported significant side effects of vomiting and patient discomfort related to the MB staining.10Dave U Shousha S Westaby D. Methylene blue staining: is it really useful in Barrett's esophagus?.Gastrointest Endosc. 2001; 53: 333-335Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar Of the two studies in which MB staining versus random biopsies were performed in the same patients, one8Canto MIF Setrakian S Willis J Chak A Petras R Powe NR et al.Methylene blue-directed biopsies improve detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2000; 51: 560-568Abstract Full Text Full Text PDF PubMed Scopus (257) Google Scholar found a significant decrement in the number of specimens needed to make a diagnosis of SIM and dysplasia (9 vs. 14, respectively) whereas the other5Wo JM Ray MB Mayfield-Stokes S Al-Sabbagh G Gebrail F Slone SP et al.Comparison of methylene blue-directed biopsies and conventional biopsies in the detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2001; 54: 294-301Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar found no difference in the number of specimens taken (14 vs. 12, respectively)8Canto MIF Setrakian S Willis J Chak A Petras R Powe NR et al.Methylene blue-directed biopsies improve detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2000; 51: 560-568Abstract Full Text Full Text PDF PubMed Scopus (257) Google Scholar or in the detection rate for SIM and dysplasia.5Wo JM Ray MB Mayfield-Stokes S Al-Sabbagh G Gebrail F Slone SP et al.Comparison of methylene blue-directed biopsies and conventional biopsies in the detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2001; 54: 294-301Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar Unfortunately, the biopsy protocols used in these studies differed significantly, with the result that more specimens were taken in the study of Wo et al. In the 2 studies in which more than 20 patients with SSBE were included, MB staining improved SIM detection in one,6Sharma P Topalovski M Mayo MS Weston AP. Methylene blue chromoendoscopy for detection of short segment Barrett's esophagus.Gastrointest Endosc. 2001; 54: 289-293Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar but was less effective than random biopsy in the other.5Wo JM Ray MB Mayfield-Stokes S Al-Sabbagh G Gebrail F Slone SP et al.Comparison of methylene blue-directed biopsies and conventional biopsies in the detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2001; 54: 294-301Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar When all published reports are considered together, it is clear that the actual technique of MB staining varies widely; it even differs in studies performed by the same investigator. In the first study, Canto et al.7Canto MI Setrakian S Petras RE Blades E Chak A Sivak Jr., MV Methylene blue selectively stains intestinal metaplasia in Barrett's esophagus.Gastrointest Endosc. 1996; 44: 1-7Abstract Full Text Full Text PDF PubMed Scopus (270) Google Scholar used 8 to 16 mL of acetylcysteine for Barrett's segments from 0 to 6 cm in length, and 16 to 32 mL for segments greater than 6 cm long, whereas in the second study, 16 mL was used for every 5 cm (3 mL/cm). Likewise, 10 to 15 mL of MB was used for 0- to 6-cm segments and 15 to 20 mL for segments greater than 6 cm, whereas in the second study,8Canto MIF Setrakian S Willis J Chak A Petras R Powe NR et al.Methylene blue-directed biopsies improve detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2000; 51: 560-568Abstract Full Text Full Text PDF PubMed Scopus (257) Google Scholar 20 mL of MB was administered for each 5-cm segment (4 mL/cm). It seems that with more experience, Canto et al. began using more acetylcysteine and MB. Kiesslich et al.9Kiesslich R Han M Herrmann G Jung M. Screening for specialized columnar epithelium with methylene blue: chromoendoscopy in patients with Barrett's esophagus and a normal control group.Gastrointest Endosc. 2001; 53: 47-52Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar used 10 mL/cm of acetylcysteine, Dave et al.10Dave U Shousha S Westaby D. Methylene blue staining: is it really useful in Barrett's esophagus?.Gastrointest Endosc. 2001; 53: 333-335Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar 3.3 mL/cm, and Sharma et al.6Sharma P Topalovski M Mayo MS Weston AP. Methylene blue chromoendoscopy for detection of short segment Barrett's esophagus.Gastrointest Endosc. 2001; 54: 289-293Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar did not state how much was used. Although all investigators used 0.5% MB and sprayed 1.5 to 3 mL/cm, Kiesslich et al.9Kiesslich R Han M Herrmann G Jung M. Screening for specialized columnar epithelium with methylene blue: chromoendoscopy in patients with Barrett's esophagus and a normal control group.Gastrointest Endosc. 2001; 53: 47-52Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar used 1% MB but sprayed only 1 mL/cm. Initially, Canto et al. sprayed MB immediately after the application of acetylcyteine, followed by an immediate wash and subsequent biopsies. However, in their second study, the procedure was modified and a 2-minute dwell time was added after the acetylcysteine and MB application, a technique used by most investigators.8Canto MIF Setrakian S Willis J Chak A Petras R Powe NR et al.Methylene blue-directed biopsies improve detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2000; 51: 560-568Abstract Full Text Full Text PDF PubMed Scopus (257) Google Scholar The volume of wash used has varied between 10 and 30 mL/cm with either water or normal saline solution, with one investigator7Canto MI Setrakian S Petras RE Blades E Chak A Sivak Jr., MV Methylene blue selectively stains intestinal metaplasia in Barrett's esophagus.Gastrointest Endosc. 1996; 44: 1-7Abstract Full Text Full Text PDF PubMed Scopus (270) Google Scholar stating that an excessively vigorous washing may diminish the MB stain, but a later study indicated that vigorous washing is important for good results.8Canto MIF Setrakian S Willis J Chak A Petras R Powe NR et al.Methylene blue-directed biopsies improve detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2000; 51: 560-568Abstract Full Text Full Text PDF PubMed Scopus (257) Google Scholar It would be highly desirable to finally have a reproducible, reliable, standardized staining technique to facilitate collation and comparison of data from future research. Further studies are needed to understand the mechanism and kinetics of acetylcysteine proteolysis and MB staining, the location and duration of MB staining in the cell, and the reproducibility of the staining pattern within areas of SIM. Another aspect of MB staining that causes confusion is the various shades of MB stain that remain after a wash. The shades of blue color vary from markedly dark to extremely light blue. There may be large patches of blue stain, resulting in difficulty in determining how many specimens should be taken. Also, in stained areas, it is necessary to decide whether to obtain specimens from only the dark blue area. There is no clear understanding of the relationship between MB staining intensity and the presence of SIM, dysplasia, and cancer. Therefore, the tendency is to take specimens from tissues with all shades of staining. But if many specimens are taken, then the rationale for MB staining becomes moot. When there is diffuse staining and the color intensity is similar throughout, the procedure takes on an unguided random format because there is no color difference that can be used to direct biopsies. None of the published studies have included data on the yield of SIM or dysplasia when taking specimens of “light blue” stained areas, although in a preliminary report, Canto et al.15Canto M Setrakians S Willis J Petras R Chak A Sivak Jr., MV Methylene blue staining of dysplastic and non-dysplastic Barrett's: an in vivo and ex vivo study.Gastrointest Endosc. 1996; 43 ([abstract]): AB331Google Scholar noted that lighter MB stains in ex vivo tissue could differentiate between cardiac or fundic-type mucosa. In their initial study, these investigators obtained specimens from only the dark MB-stained areas, which yielded SIM, dysplasia, and intramucosal cancer.7Canto MI Setrakian S Petras RE Blades E Chak A Sivak Jr., MV Methylene blue selectively stains intestinal metaplasia in Barrett's esophagus.Gastrointest Endosc. 1996; 44: 1-7Abstract Full Text Full Text PDF PubMed Scopus (270) Google Scholar However, in another publication, it was suggested that lighter-stained areas within darker-stained regions will have a higher yield of dysplasia and cancer because of the increased nuclear-to-cytoplasmic ratio and fewer goblet cells to absorb less MB.8Canto MIF Setrakian S Willis J Chak A Petras R Powe NR et al.Methylene blue-directed biopsies improve detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2000; 51: 560-568Abstract Full Text Full Text PDF PubMed Scopus (257) Google Scholar To add to the confusion, Sharma et al.,6Sharma P Topalovski M Mayo MS Weston AP. Methylene blue chromoendoscopy for detection of short segment Barrett's esophagus.Gastrointest Endosc. 2001; 54: 289-293Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar although agreeing with this hypothesis, reported that all dysplastic specimens found in their study came from darkly stained MB tissue. Until there are studies that histologically localize the MB in differing cell types, the significance of variations in the intensity of MB staining will remain unresolved. One of the major differences in the various studies of MB chromoendoscopy is the way the data are expressed. Some investigators base the sensitivity or detection rate on number of biopsy specimens,5Wo JM Ray MB Mayfield-Stokes S Al-Sabbagh G Gebrail F Slone SP et al.Comparison of methylene blue-directed biopsies and conventional biopsies in the detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2001; 54: 294-301Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar, 6Sharma P Topalovski M Mayo MS Weston AP. Methylene blue chromoendoscopy for detection of short segment Barrett's esophagus.Gastrointest Endosc. 2001; 54: 289-293Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar, 10Dave U Shousha S Westaby D. Methylene blue staining: is it really useful in Barrett's esophagus?.Gastrointest Endosc. 2001; 53: 333-335Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar whereas others base it on number of patients. In the per-specimen calculation, sensitivity is expressed as a proportion of SIM specimens that are detected by MB. This method gives the best indication of how good MB staining is with respect to the identification of SIM. In contrast, when sensitivity is expressed in terms of number of patients, even if 2 of 10 stained specimens are positive for SIM, the patient is included as a test positive in the sensitivity calculation.6Sharma P Topalovski M Mayo MS Weston AP. Methylene blue chromoendoscopy for detection of short segment Barrett's esophagus.Gastrointest Endosc. 2001; 54: 289-293Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar, 7Canto MI Setrakian S Petras RE Blades E Chak A Sivak Jr., MV Methylene blue selectively stains intestinal metaplasia in Barrett's esophagus.Gastrointest Endosc. 1996; 44: 1-7Abstract Full Text Full Text PDF PubMed Scopus (270) Google Scholar, 8Canto MIF Setrakian S Willis J Chak A Petras R Powe NR et al.Methylene blue-directed biopsies improve detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2000; 51: 560-568Abstract Full Text Full Text PDF PubMed Scopus (257) Google Scholar Canto et al.7Canto MI Setrakian S Petras RE Blades E Chak A Sivak Jr., MV Methylene blue selectively stains intestinal metaplasia in Barrett's esophagus.Gastrointest Endosc. 1996; 44: 1-7Abstract Full Text Full Text PDF PubMed Scopus (270) Google Scholar initially noted that the sensitivity and specificity of MB-stained specimens for identifying SIM to be 95% and 97%, respectively. No other investigator has been able to match these results. In this issue of the Journal, Wo et al.5Wo JM Ray MB Mayfield-Stokes S Al-Sabbagh G Gebrail F Slone SP et al.Comparison of methylene blue-directed biopsies and conventional biopsies in the detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2001; 54: 294-301Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar noted that 75% of MB-stained specimens versus 72% of unstained specimens contained SIM. Likewise, Dave et al.10Dave U Shousha S Westaby D. Methylene blue staining: is it really useful in Barrett's esophagus?.Gastrointest Endosc. 2001; 53: 333-335Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar found the sensitivity and specificity in identifying SIM of stained versus unstained specimens was only 57% versus 33%, respectively. On the other hand, if the number of patients who had SIM detected by MB staining is considered, the results are better and statistically more likely to be significant. The recently published larger study of Canto et al.8Canto MIF Setrakian S Willis J Chak A Petras R Powe NR et al.Methylene blue-directed biopsies improve detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2000; 51: 560-568Abstract Full Text Full Text PDF PubMed Scopus (257) Google Scholar expressed data in terms of the percentage of patients with MB staining versus random biopsies (91% vs. 69%, p = 0.001), respectively. Sharma et al.,6Sharma P Topalovski M Mayo MS Weston AP. Methylene blue chromoendoscopy for detection of short segment Barrett's esophagus.Gastrointest Endosc. 2001; 54: 289-293Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar studying SSBE, noted that the SIM detection rate was 61% for patients in the MB staining group versus 42% for the random biopsy specimen patient group (p = 0.02). Kiesslich et al.9Kiesslich R Han M Herrmann G Jung M. Screening for specialized columnar epithelium with methylene blue: chromoendoscopy in patients with Barrett's esophagus and a normal control group.Gastrointest Endosc. 2001; 53: 47-52Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar noted a specificity of 98% for patients in whom MB staining was positive. Of critical importance is the diagnosis of dysplasia and cancer. Canto et al7Canto MI Setrakian S Petras RE Blades E Chak A Sivak Jr., MV Methylene blue selectively stains intestinal metaplasia in Barrett's esophagus.Gastrointest Endosc. 1996; 44: 1-7Abstract Full Text Full Text PDF PubMed Scopus (270) Google Scholar initially reported that dysplasia was found in 16% of stained specimens and in only 1.1% of unstained. Later, this group noted that the diagnostic rate for MB staining for patients with dysplasia and cancer was significantly greater compared with random biopsy (44% vs. 28%, respectively).8Canto MIF Setrakian S Willis J Chak A Petras R Powe NR et al.Methylene blue-directed biopsies improve detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2000; 51: 560-568Abstract Full Text Full Text PDF PubMed Scopus (257) Google Scholar With MB staining, a total of 7 more patients were found by using the MB technique. In patients with SSBE, Sharma et al.6Sharma P Topalovski M Mayo MS Weston AP. Methylene blue chromoendoscopy for detection of short segment Barrett's esophagus.Gastrointest Endosc. 2001; 54: 289-293Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar found 6 patients who were MB positive with dysplasia, although no mention was made of the control group. Of interest, Wo et al5Wo JM Ray MB Mayfield-Stokes S Al-Sabbagh G Gebrail F Slone SP et al.Comparison of methylene blue-directed biopsies and conventional biopsies in the detection of intestinal metaplasia and dysplasia in Barrett's esophagus.Gastrointest Endosc. 2001; 54: 294-301Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar noted no difference in the proportion of specimens containing dysplasia by MB staining or random specimens (20% vs. 18%, respectively). Likewise, within specimens obtained after MB staining, positively stained specimens were no better with regard to the diagnosis of dysplasia than negatively stained specimens. So how can all this information be put together and sense be made of it? Is MB staining worth doing? From our experience, MB staining is not as simple to perform as initially anticipated. It requires additional reagents and supplies and consumes approximately 15 to 20 minutes to prepare and dilute MB and acetylcysteine before the procedure. Special spray catheters must be purchased to apply the MB and acetylcysteine to the esophagus. Also, a second person is required to assist the endoscopist in addition to the usual support staff. The risk of aspiration may be much greater in patients with LSBE because of the large volume of washing fluid used. All mucosal surfaces must be evenly sprayed with acetylcysteine with appropriate dwell time, and adequate volume of MB must be applied to all surfaces. When the procedure was first performed, the focal nature of the stain and the many shades of blue that made it difficult to determine where and how many specimens should be taken were surprising. Interestingly, our data indicate that with MB staining, 50% fewer specimens are required to identify SIM and dysplasia as compared with random biopsies.16Horwhat JD Ramos F Colina R Maydonovitch CL Wong RKH. A prospective, controlled study to compare the diagnostic yield of methylene blue directed biopsies versus 4-quadrant random biopsies to identify the presence and grade of dysplasia in Barrett's esophagus.Gastroenterology. 1999; 116 ([abstract]): A189Google Scholar, 17Horwhat JD Ramos F Colina R Maydonovitch CL Wong RKH. Does dark staining or non-staining predict the presence of dysplasia in patients with Barrett's esophagus undergoing methylene blue directed biopsies.Am J Gastroenterol. 1999; 94 ([abstract]): A80Crossref PubMed Scopus (79) Google Scholar However, MB stained biopsies do not always identify SIM or dysplasia because 25% to 45% of MB-stained specimens contain neither SIM nor dysplasia. Overall, these data indicate that (1) there are methodologic differences between reported studies, and (2) MB staining is not as sensitive and specific in identifying SIM or dysplasia when examining individual MB-stained specimens; that is, there are significant numbers of false-positive and false-negative specimens. However, (3) when considering the overall diagnostic yield within patients, the technique of MB staining is equal to random biopsies in identifying some focus of SIM or dysplasia and may require fewer specimens. Finally, (4) the technique is more time consuming and may be associated with increased morbidity. Refinements and standardization concerning the technique will allow for equal comparisons.